Most of what is currently known about human brain aging based on neuroimaging studies is only inferred from cross-sectional data. There have been very few longitudinal imaging studies and relatively little focus on middle age, despite the fact that several studies suggest that midlife is a key inflection point for brain aging. Even less is known about the extent of genetic and environmental influences on brain aging within a longitudinal design. The proposed study will fill these critical knowledge gaps with the first 5-year follow-up of The VETSA Longitudinal MRI Twin Study of Aging (VMRI). (VETSA is an acronym for Vietnam Era Twin Study of Aging.) We enrolled 607 twins and obtained 515 (85 percent) analyzable scans in the baseline VMRI with comprehensive structural and diffusion tensor imaging. We propose to replace some subjects and acquire 600 scans in this first follow-up study. Twins were 51-60 years old at the baseline assessment and will be 56-65 in the proposed VMRI follow-up. Given the large sample and narrow age range, we will have maximal power to examine within-person change and individual differences in change during this key transition period. Specific aims are to: 1) add a longitudinal imaging component to our unique normative database; 2) determine genetic and environmental influences on brain structure changes over time; 3) examine the relationship of APOE genotype to changes in brain structure over time; 4) elucidate biomedical and other risk factors related to changes in brain structure over time, and examine their shared genetic and environmental underpinnings; 5) provide detailed characterization of white matter changes with age; and 6) examine brain function and integration of regional brain activity by studying regional perfusion (via arterial spin labeling) and the default network (via functional MRI). Potential risk/protective factors are available in the extensive cognitive, biomedical, and psychosocial phenotypes collected in the parent project. In keeping with a 2-factor model, we propose that normal aging primarily affects frontal-striatal systems whereas the medial temporal system is more strongly affected in Alzheimer's disease. White matter integrity is hypothesized to underlie functional connectivity, but there exists only cross-sectional evidence. We hypothesize that baseline white matter integrity and change in white matter integrity over time will predict the strength of activity correlations between major components of the default network. Based on our current findings, we also predict greater cortical thinning in prefrontal regions in APOE-54 carrier. The proposed project is in a unique position to identify early predictors of successful or pathological brain aging and delineate their relative underlying genetic and environmental influences. Relevance: This project creates an invaluable resource for understanding the course of brain aging beginning in midlife (an understudied period). Moreover, the possibility of identifying predictors in midlife, rather than in later life, has important public health implications with regard to intervention or prevention. PUBLIC HEALTH RELEVANCE: PROJECT NARRATIVE This project creates an invaluable resource for understanding the course of brain aging beginning in midlife (an understudied period). Moreover, the possibility of identifying predictors of brain aging in midlife, rather than in later life, has important public health implications with regard to intervention or prevention.